Collective Supervision of Topic Models for Predicting Surveys with Social Media

Adrian Benton, Michael J. Paul, Braden Hancock, Mark Dredze

AAAI 2016 pp. 2892-2898

doi:10.1609/AAAI.V30I1.10374 /aaai/2016/benton2016aaai-collective/

Abstract

This paper considers survey prediction from social media. We use topic models to correlate social media messages with survey outcomes and to provide an interpretable representation of the data. Rather than rely on fully unsupervised topic models, we use existing aggregated survey data to inform the inferred topics, a class of topic model supervision referred to as collective supervision. We introduce and explore a variety of topic model variants and provide an empirical analysis, with conclusions of the most effective models for this task.

PDF AAAI Semantic Scholar

Cite

Text

Benton et al. "Collective Supervision of Topic Models for Predicting Surveys with Social Media." AAAI Conference on Artificial Intelligence, 2016. doi:10.1609/AAAI.V30I1.10374

Markdown

[Benton et al. "Collective Supervision of Topic Models for Predicting Surveys with Social Media." AAAI Conference on Artificial Intelligence, 2016.](https://mlanthology.org/aaai/2016/benton2016aaai-collective/) doi:10.1609/AAAI.V30I1.10374

BibTeX

@inproceedings{benton2016aaai-collective,
  title     = {{Collective Supervision of Topic Models for Predicting Surveys with Social Media}},
  author    = {Benton, Adrian and Paul, Michael J. and Hancock, Braden and Dredze, Mark},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2016},
  pages     = {2892-2898},
  doi       = {10.1609/AAAI.V30I1.10374},
  url       = {https://mlanthology.org/aaai/2016/benton2016aaai-collective/}
}